Does NvBufSurfTransform API require src and dst of same memory type?

Accelerated Computing Intelligent Video Analytics DeepStream SDK
1

When I use src of cuda-unified memory and dst of cuda-pinned host memory, pipeline sometimes runs perfectly well, sometimes runs into stuck or decodes with mosaic area.

However, when I change dst also to cuda-unified memory, it always runs well.

So does it require src and dst of same memory type?

• Hardware Platform GPU
• DeepStream Version 5.1

3

There is no update from you for a period, assuming this is not an issue any more.
Hence we are closing this topic. If need further support, please open a new one.
Thanks
what’s the platform?
do you use cudaMemcpy*() ?